F_VN_PredictSampleVectorExp
Compute a vectorial prediction for a single sample. (expert function)
Can use available TwinCAT Job Tasks for executing parallel code regions.
Syntax
Definition:
FUNCTION F_VN_PredictSampleVectorExp : HRESULT
VAR_INPUT
ipRegressor : ITcVnMlModel;
ipSample : ITcUnknown;
ipPrediction : Reference To ITcVnContainer;
fNovelty : Reference To REAL;
hrPrev : HRESULT;
END_VAR
Inputs
|
Name |
Type |
Description |
|---|---|---|
|
ipRegressor |
Regressor to be used | |
|
ipSample |
Container holding a single input sample (ContainerType_Vector_REAL or ContainerType_Vector_LREAL) | |
|
ipPrediction |
Reference To ITcVnContainer |
Returns the predicted output (ContainerType_Vector_REAL or ContainerType_Vector_LREAL, depending on ipSample) |
|
fNovelty |
Reference To REAL |
Returns the degree of novelty (0.0 if a sample is completely known; > 0.0 otherwise) of the presented sample (optional, set to 0 if not required) |
|
hrPrev |
HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.) |
Return valueFurther information
The function F_VN_PredictSampleVectorExp is the expert variant of F_VN_PredictSampleVector. It contains additional parameters.
Parameter
Regression model
The previously trained regression model must be transferred to ipRegressor.
Sample
The sample container is transferred as ipSample. The container type must be either ContainerType_Vector_REAL or ContainerType_Vector_LREAL.
Prediction
The calculated prediction vector is returned as a container via the reference ipPrediction. The container type is taken from ipSamples.
Degree of novelty
The degree of novelty of the sample is returned via fNovelty.
Application
For example, the prediction vector of a sample can be calculated as follows:
hr := F_VN_PredictSampleVectorExp(
ipRegressor := ipRegressor,
ipSample := ipSample,
ipPrediction:= ipPrediction,
fNovelty := fNovelty,
hrPrev := hr);Related functions
Required License
TC3 Machine Learning Realtime Inference